Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | "Ozturk, Orgul" <odozturk@moore.sc.edu> |
To | "'statalist@hsphsun2.harvard.edu'" <statalist@hsphsun2.harvard.edu> |
Subject | RE: st: margins |
Date | Tue, 21 Jan 2014 08:07:09 -0500 |
Maarten, Thank you for your response. I have this specifications : reg logrealwagechange i.promCurrentJob##i.woman#i.govt##i.schoolinggroup $extras $industry $occupationclass if careerstage==3 , noc for this margins give me no estimates. Once I have the estimates I want to run test comparing male female estimates within each schooling, and sector by promotion status And then again I want to look at effect of promotion by looking at wage growth for promoted and nonpromoted by schooling, sector and gender. Everything runs without a problem if I ignore the nonexistence of the constant. It does not really make a difference for this case but I want to learn how it is to be estimated for future reference For the xtlogit fe I am working with xtlogit promCurrentJob i.woman##i.careerstage##i.govt##i.schoolinggroup $usualsuspects, fe I create the estimates by setting the unobserved factors equal to zero, but when I then tabulate them by career stage, sector and gender{reference group (male in early career with lowest education in private sector) ] I cannot get the varying marginal effect of early career in private sector. Even if I get these I am not perfectly sure how to interpret them. Thank you. Orgul Orgul Demet Ozturk Assistant Professor of Economics Economics Department Moore School of Business University of South Carolina https://sites.google.com/site/orguldemetozturk/ -----Original Message----- From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Maarten Buis Sent: Tuesday, January 21, 2014 3:19 AM To: statalist@hsphsun2.harvard.edu Subject: Re: st: margins --- On Mon, Jan 20, 2014 at 8:29 PM, Ozturk, Orgul wrote: > How can I produce average marginal effects after an OLS regression with no constant term. > It gives "non-estimable" I cannot reproduce your problem, the following runs without problem on Stata 12: *--------- begin example -------------- sysuse auto reg price ibn.rep78 mpg, hascons margins i.rep78 *---------- end example --------------- Can you tell us exactly what you typed? > Moreover, how can one produce AMEs using margins command after a fixed > effects logistic regression (after xtlogit, fe) You'll probably need to specify the -predict()- option of margins, how will depend on what _exactly_ you want. -- Maarten --------------------------------- Maarten L. Buis WZB Reichpietschufer 50 10785 Berlin Germany http://www.maartenbuis.nl --------------------------------- * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/faqs/resources/statalist-faq/ * http://www.ats.ucla.edu/stat/stata/ * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/faqs/resources/statalist-faq/ * http://www.ats.ucla.edu/stat/stata/